Ardupilot Rpanion iperf网络性能测试

Ardupilot Rpanion iperf网络性能测试

  • 1. 源由
  • 2. 分析
  • 3. 安装
  • 4. 测试
    • 4.1 第一次测试
      • 4.1.1 iperf测试参数A
        • 4.1.1.1 测试链路1
        • 4.1.1.2 测试链路2
        • 4.1.1.3 测试链路3
      • 4.1.2 iperf测试参数B - 测试链路3
        • 4.1.2.1 测试数据
        • 4.1.2.2 数据简单分析
        • 4.1.2.3 数据深入分析
        • 4.1.2.4 模拟测试网络
    • 4.2 第二次测试
      • 4.2.1 测试数据
      • 4.2.2 数据分析
      • 4.2.3 网络延迟分析
      • 4.2.4 视频效果
    • 4.3 第三次测试
      • 4.3.1 测试数据
      • 4.3.2 数据分析
      • 4.3.3 网络延迟分析
      • 4.3.4 视频效果
  • 5. 总结
  • 6. 参考资料
  • 7. ZeroTier优化办法 - Moon节点
    • 7.1 服务端配置
    • 7.2 客户端配置
  • 8. DJI RC系统性能

1. 源由

网络性能测试主要目的是:

  • 评估网络健康状况:网络性能测试可以帮助评估网络的健康状况,包括带宽、延迟、丢包率等指标。通过了解网络的实际性能,可以及时发现并解决网络瓶颈和问题,确保网络的正常运行。

  • 优化网络性能:通过网络性能测试,可以确定网络中存在的瓶颈和性能问题,并采取相应的优化措施。这包括调整网络设备配置、优化网络拓扑、升级网络硬件等,以提高网络的吞吐量、稳定性和可靠性。

  • 规划网络扩展和升级:网络性能测试可以帮助企业规划网络的扩展和升级,以满足不断增长的业务需求。通过了解当前网络的性能状况和未来的需求,可以更好地规划网络架构、设备采购和部署策略。

  • 保障用户体验:良好的网络性能是保障用户体验的关键。网络性能测试可以帮助确保网络对用户提供良好的服务质量,包括快速的响应时间、稳定的连接和高质量的数据传输。

  • 提高网络安全:网络性能测试还可以帮助发现和解决网络安全漏洞和风险。通过测试网络的性能和安全性,可以识别潜在的安全威胁,并采取相应的防护措施,提高网络的安全性和可靠性。

最近要做的是针对视频、电传链路的评估,其在VPN/4G-LTE下性能到底如何?

  • 《ArduPilot开源代码之CompanionComputer上天计划》
  • 《Ardupilot开源代码之Rover上路计划》

2. 分析

在进行网络性能测试时,其主要步骤如下:

  • 清晰的测试目标:确定您要测试的网络性能指标,例如带宽、延迟、丢包率等。确保您的测试目标清晰明确,并且测试过程中能够量化和记录这些指标。

  • 选择适当的测试工具:根据您的测试需求选择合适的工具。一些常用的网络性能测试工具包括iperf、Wireshark、PingPlotter等。确保您选择的工具能够提供您需要的功能和数据输出。

  • 测试环境的准备:在进行测试之前,确保测试环境的稳定性和一致性。避免在网络负载较高的时段进行测试,以免影响测试结果。另外,确保测试设备和网络连接的稳定性,以避免测试过程中出现干扰或故障。

  • 数据采集和分析:在测试过程中,及时记录测试数据,并进行分析和比较。这些数据可以帮助您评估网络性能,并识别潜在的问题和瓶颈。确保您使用的工具能够提供详细的测试结果和分析功能。

  • 安全性和合规性:在进行网络性能测试时,确保您的测试活动不会对网络安全造成威胁,并遵守相关的法律法规和政策要求。在测试之前,与相关部门或团队沟通,确保您的测试活动符合组织的安全和合规标准。

  • 测试结果的验证和确认:在完成测试之后,验证和确认测试结果的准确性和可靠性。如果可能,进行多次测试以确保结果的一致性,并与预期的性能指标进行比较。

综上所述,网络性能测试需要仔细的规划和准备工作,以确保测试结果的准确性和可信度。通过遵循上述注意事项,您可以更好地评估和优化网络性能,提高网络的稳定性和效率。

3. 安装

  • Jetson Orin Nano
$ sudo apt-get install iperf
$ iperf --version
iperf version 2.1.5 (3 December 2021) pthreads
  • Raspberrry Pi 3B+
$ sudo apt-get install iperf
$ iperf --version
iperf version 2.0.14a (2 October 2020) pthreads
$ iperf --version
iperf version 2.1.5 (3 December 2021) pthreads
daniel@daniel-nvidia:~$ iperf -h
Usage: iperf [-s|-c host] [options]iperf [-h|--help] [-v|--version]Client/Server:-b, --bandwidth #[kmgKMG | pps]  bandwidth to read/send at in bits/sec or packets/sec-e, --enhanced    use enhanced reporting giving more tcp/udp and traffic information-f, --format    [kmgKMG]   format to report: Kbits, Mbits, KBytes, MBytes--hide-ips           hide ip addresses and host names within outputs-i, --interval  #        seconds between periodic bandwidth reports-l, --len       #[kmKM]    length of buffer in bytes to read or write (Defaults: TCP=128K, v4 UDP=1470, v6 UDP=1450)-m, --print_mss          print TCP maximum segment size (MTU - TCP/IP header)-o, --output    <filename> output the report or error message to this specified file-p, --port      #        client/server port to listen/send on and to connect--permit-key         permit key to be used to verify client and server (TCP only)--sum-only           output sum only reports-u, --udp                use UDP rather than TCP-w, --window    #[KM]    TCP window size (socket buffer size)-z, --realtime           request realtime scheduler-B, --bind <host>[:<port>][%<dev>] bind to <host>, ip addr (including multicast address) and optional port and device-C, --compatibility      for use with older versions does not sent extra msgs-M, --mss       #        set TCP maximum segment size (MTU - 40 bytes)-N, --nodelay            set TCP no delay, disabling Nagle's Algorithm-S, --tos       #        set the socket's IP_TOS (byte) field-Z, --tcp-congestion <algo>  set TCP congestion control algorithm (Linux only)Server specific:-p, --port      #[-#]    server port(s) to listen on/connect to-s, --server             run in server mode-1, --singleclient       run one server at a time--histograms         enable latency histograms--permit-key-timeout set the timeout for a permit key in seconds--tcp-rx-window-clamp set the TCP receive window clamp size in bytes--tap-dev   #[<dev>] use TAP device to receive at L2 layer-t, --time      #        time in seconds to listen for new connections as well as to receive traffic (default not set)--udp-histogram #,#  enable UDP latency histogram(s) with bin width and count, e.g. 1,1000=1(ms),1000(bins)-B, --bind <ip>[%<dev>]  bind to multicast address and optional device-U, --single_udp         run in single threaded UDP mode--sum-dstip          sum traffic threads based upon destination ip address (default is src ip)-D, --daemon             run the server as a daemon-V, --ipv6_domain        Enable IPv6 reception by setting the domain and socket to AF_INET6 (Can receive on both IPv4 and IPv6)Client specific:-c, --client    <host>   run in client mode, connecting to <host>--connect-only       run a connect only test--connect-retries #  number of times to retry tcp connect-d, --dualtest           Do a bidirectional test simultaneously (multiple sockets)--fq-rate #[kmgKMG]  bandwidth to socket pacing--full-duplex        run full duplex test using same socket--ipg                set the the interpacket gap (milliseconds) for packets within an isochronous frame--isochronous <frames-per-second>:<mean>,<stddev> send traffic in bursts (frames - emulate video traffic)--incr-dstip         Increment the destination ip with parallel (-P) traffic threads--incr-dstport       Increment the destination port with parallel (-P) traffic threads--incr-srcip         Increment the source ip with parallel (-P) traffic threads--incr-srcport       Increment the source port with parallel (-P) traffic threads--local-only         Set don't route on socket--near-congestion=[w] Use a weighted write delay per the sampled TCP RTT (experimental)--no-connect-sync    No sychronization after connect when -P or parallel traffic threads--no-udp-fin         No final server to client stats at end of UDP test-n, --num       #[kmgKMG]    number of bytes to transmit (instead of -t)-r, --tradeoff           Do a fullduplexectional test individually--tcp-write-prefetch set the socket's TCP_NOTSENT_LOWAT value in bytes and use event based writes-t, --time      #        time in seconds to transmit for (default 10 secs)--trip-times         enable end to end measurements (requires client and server clock sync)--txdelay-time       time in seconds to hold back after connect and before first write--txstart-time       unix epoch time to schedule first write and start traffic-B, --bind [<ip> | <ip:port>] bind ip (and optional port) from which to source traffic-F, --fileinput <name>   input the data to be transmitted from a file-H, --ssm-host <ip>      set the SSM source, use with -B for (S,G)-I, --stdin              input the data to be transmitted from stdin-L, --listenport #       port to receive fullduplexectional tests back on-P, --parallel  #        number of parallel client threads to run-R, --reverse            reverse the test (client receives, server sends)-S, --tos                IP DSCP or tos settings-T, --ttl       #        time-to-live, for multicast (default 1)-V, --ipv6_domain        Set the domain to IPv6 (send packets over IPv6)-X, --peer-detect        perform server version detection and version exchangeMiscellaneous:-x, --reportexclude [CDMSV]   exclude C(connection) D(data) M(multicast) S(settings) V(server) reports-y, --reportstyle C      report as a Comma-Separated Values-h, --help               print this message and quit-v, --version            print version information and quit[kmgKMG] Indicates options that support a k,m,g,K,M or G suffix
Lowercase format characters are 10^3 based and uppercase are 2^n based
(e.g. 1k = 1000, 1K = 1024, 1m = 1,000,000 and 1M = 1,048,576)The TCP window size option can be set by the environment variable
TCP_WINDOW_SIZE. Most other options can be set by an environment variable
IPERF_<long option name>, such as IPERF_BANDWIDTH.Source at <http://sourceforge.net/projects/iperf2/>
Report bugs to <iperf-users@lists.sourceforge.net>

4. 测试

  • 测试指标:带宽、延迟、丢包率、稳定性。其他的可靠性、安全性等暂不考虑。
  • 测试工具:iperf

4.1 第一次测试

目前,本地已有的测试环境的【网络拓扑A】,如下所示:
在这里插入图片描述

4.1.1 iperf测试参数A

  • 服务端:iperf -u -s
  • 客户端:iperf -u -T 10 -b [1M] -c [server address]
4.1.1.1 测试链路1
  • 局域网测试:Pi --> Hub --> Jetson
  • 服务端IP地址:192.168.1.19 (Jetson)
  • 客户端IP地址:192.168.1.201 (Pi)
$ iperf -u -T 10 -b 1M -c 192.168.1.19
------------------------------------------------------------
Client connecting to 192.168.1.19, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 192.168.1.201 port 49022 connected with 192.168.1.19 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0153 sec  1.25 MBytes  1.05 Mbits/sec
[  3] Sent 896 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0116 sec  1.25 MBytes  1.05 Mbits/sec   3.065 ms    0/  895 (0%)$ iperf -u -T 10 -b 2M -c 192.168.1.19
------------------------------------------------------------
Client connecting to 192.168.1.19, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 192.168.1.201 port 46959 connected with 192.168.1.19 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0098 sec  2.50 MBytes  2.10 Mbits/sec
[  3] Sent 1788 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-9.9998 sec  2.50 MBytes  2.10 Mbits/sec   2.513 ms    0/ 1787 (0%)$ iperf -u -T 10 -b 3M -c 192.168.1.19
------------------------------------------------------------
Client connecting to 192.168.1.19, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 192.168.1.201 port 44713 connected with 192.168.1.19 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0079 sec  3.75 MBytes  3.15 Mbits/sec
[  3] Sent 2680 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0206 sec  3.75 MBytes  3.14 Mbits/sec   2.352 ms    1/ 2679 (0.037%)
  • 1M带宽测试:
    带宽:1.05 Mbits/sec
    丢包率:0%
    Jitter:3.065 ms

  • 2M带宽测试:
    带宽:2.10 Mbits/sec
    丢包率:0%
    Jitter:2.513 ms

  • 3M带宽测试:
    带宽:3.14 Mbits/sec
    丢包率:0.037%
    Jitter:2.352 ms

4.1.1.2 测试链路2
  • 预设条件:4G LTE SIM卡移除
  • VPN测试:Pi --> Hub --> Modem --> Server --> Modem --> Hub --> AP --> Jetson
  • 服务端IP地址:172.24.42.46 (Jetson)
  • 客户端IP地址:172.24.161.33 (Pi)
$ iperf -u -T 10 -b 1M -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33 port 43642 connected with 172.24.42.46 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0153 sec  1.25 MBytes  1.05 Mbits/sec
[  3] Sent 896 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-9.9983 sec  1.25 MBytes  1.05 Mbits/sec   3.425 ms    0/  895 (0%)$ iperf -u -T 10 -b 2M -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33 port 43610 connected with 172.24.42.46 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0098 sec  2.50 MBytes  2.10 Mbits/sec
[  3] Sent 1788 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0097 sec  2.50 MBytes  2.10 Mbits/sec   3.465 ms    0/ 1787 (0%)$ iperf -u -T 10 -b 3M -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33 port 37048 connected with 172.24.42.46 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0041 sec  3.75 MBytes  3.15 Mbits/sec
[  3] Sent 2679 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0336 sec  3.75 MBytes  3.14 Mbits/sec   4.807 ms    0/ 2678 (0%)
  • 1M带宽测试:
    带宽:1.05 Mbits/sec
    丢包率:0%
    Jitter:3.425 ms

  • 2M带宽测试:
    带宽:2.10 Mbits/sec
    丢包率:0%
    Jitter:3.465 ms

  • 3M带宽测试:
    带宽:3.14 Mbits/sec
    丢包率:0%
    Jitter:4.807 ms

4.1.1.3 测试链路3
  • 预设条件:有线IP网络物理断开
  • VPN测试:Pi --> 4G LTE --> BTS --> Server --> Modem --> Hub --> AP --> Jetson
  • 服务端IP地址:172.24.42.46 (Jetson)
  • 客户端IP地址:172.24.161.33 (Pi)
$ iperf -u -T 10 -b 1M -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33 port 34869 connected with 172.24.42.46 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0154 sec  1.25 MBytes  1.05 Mbits/sec
[  3] Sent 896 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-9.9907 sec  1.25 MBytes  1.05 Mbits/sec   8.210 ms    0/  895 (0%)$ iperf -u -T 10 -b 2M -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33 port 35665 connected with 172.24.42.46 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0098 sec  2.50 MBytes  2.10 Mbits/sec
[  3] Sent 1788 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-9.9933 sec  2.50 MBytes  2.10 Mbits/sec   7.350 ms    0/ 1787 (0%)$ iperf -u -T 10 -b 3M -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33 port 43883 connected with 172.24.42.46 port 5001
[ ID] Interval       Transfer     Bandwidth
[  3] 0.0000-10.0079 sec  3.75 MBytes  3.15 Mbits/sec
[  3] Sent 2680 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0150 sec  3.75 MBytes  3.14 Mbits/sec   5.268 ms    0/ 2679 (0%)
  • 1M带宽测试:
    带宽:1.05 Mbits/sec
    丢包率:0%
    Jitter:8.210 ms

  • 2M带宽测试:
    带宽:2.10 Mbits/sec
    丢包率:0%
    Jitter:7.350 ms

  • 3M带宽测试:
    带宽:3.14 Mbits/sec
    丢包率:0%
    Jitter:5.268 ms

4.1.2 iperf测试参数B - 测试链路3

为什么要引入测试参数B?主要原因是:

  1. 4G LTE视频发现花屏+卡顿现象
  2. 测试参数A给出的1/2/3链路状况都是非常理想的
    上述两个现象与实际结果发生了严重的不一致问题,太困惑了。
  • 4G LTE networks results in significant screen freezing and stuttering #225

怀疑:是否是iperf测试参数问题导致测试结果不准确。因此,调整测试参数如下:

  • 服务端:iperf -u -s
  • 客户端:iperf -u -T 10 -b [1M] -l 20 -e -i 1 -t 60 -c [server address]

预设条件:有线IP网络物理断开
VPN测试:Pi --> 4G LTE --> BTS --> Server --> Modem --> Hub --> AP --> Jetson
服务端IP地址:172.24.42.46 (Jetson)
客户端IP地址:172.24.161.33 (Pi)

4.1.2.1 测试数据
$ iperf -u -T 10 -b 2M -l 20 -e -i 1 -t 60 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 950 (1 flows)
Write buffer size: 20.0 Byte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 57316 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec   256 KBytes  2.10 Mbits/sec  13104/0    13105 pps
[  3] 1.0000-2.0000 sec   256 KBytes  2.10 Mbits/sec  13111/0    13112 pps
[  3] 2.0000-3.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13106 pps
[  3] 3.0000-4.0000 sec   256 KBytes  2.10 Mbits/sec  13106/0    13108 pps
[  3] 4.0000-5.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13108 pps
[  3] 5.0000-6.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13105 pps
[  3] 6.0000-7.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13107 pps
[  3] 7.0000-8.0000 sec   256 KBytes  2.10 Mbits/sec  13109/0    13108 pps
[  3] 8.0000-9.0000 sec   256 KBytes  2.10 Mbits/sec  13103/0    13103 pps
[  3] 9.0000-10.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13112 pps
[  3] 10.0000-11.0000 sec   256 KBytes  2.10 Mbits/sec  13111/0    13106 pps
[  3] 11.0000-12.0000 sec   256 KBytes  2.10 Mbits/sec  13106/0    13108 pps
[  3] 12.0000-13.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13107 pps
[  3] 13.0000-14.0000 sec   256 KBytes  2.10 Mbits/sec  13105/0    13107 pps
[  3] 14.0000-15.0000 sec   256 KBytes  2.10 Mbits/sec  13105/0    13102 pps
[  3] 15.0000-16.0000 sec   256 KBytes  2.10 Mbits/sec  13112/0    13113 pps
[  3] 16.0000-17.0000 sec   256 KBytes  2.10 Mbits/sec  13104/0    13104 pps
[  3] 17.0000-18.0000 sec   256 KBytes  2.10 Mbits/sec  13109/0    13111 pps
[  3] 18.0000-19.0000 sec   256 KBytes  2.10 Mbits/sec  13109/0    13107 pps
[  3] 19.0000-20.0000 sec   256 KBytes  2.10 Mbits/sec  13105/0    13107 pps
[  3] 20.0000-21.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13105 pps
[  3] 21.0000-22.0000 sec   256 KBytes  2.10 Mbits/sec  13105/0    13109 pps
[  3] 22.0000-23.0000 sec   256 KBytes  2.10 Mbits/sec  13097/0    13107 pps
[  3] 23.0000-24.0000 sec   256 KBytes  2.10 Mbits/sec  13120/0    13107 pps
[  3] 24.0000-25.0000 sec   256 KBytes  2.10 Mbits/sec  13106/0    13105 pps
[  3] 25.0000-26.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13110 pps
[  3] 26.0000-27.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13105 pps
[  3] 27.0000-28.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13109 pps
[  3] 28.0000-29.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13105 pps
[  3] 29.0000-30.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13107 pps
[  3] 30.0000-31.0000 sec   256 KBytes  2.10 Mbits/sec  13106/0    13106 pps
[  3] 31.0000-32.0000 sec   256 KBytes  2.10 Mbits/sec  13109/0    13110 pps
[  3] 32.0000-33.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13108 pps
[  3] 33.0000-34.0000 sec   256 KBytes  2.10 Mbits/sec  13104/0    13107 pps
[  3] 34.0000-35.0000 sec   256 KBytes  2.10 Mbits/sec  13109/0    13105 pps
[  3] 35.0000-36.0000 sec   256 KBytes  2.10 Mbits/sec  13109/0    13109 pps
[  3] 36.0000-37.0000 sec   256 KBytes  2.10 Mbits/sec  13106/0    13108 pps
[  3] 37.0000-38.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13105 pps
[  3] 38.0000-39.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13109 pps
[  3] 39.0000-40.0000 sec   256 KBytes  2.10 Mbits/sec  13106/0    13107 pps
[  3] 40.0000-41.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13107 pps
[  3] 41.0000-42.0000 sec   256 KBytes  2.10 Mbits/sec  13109/0    13107 pps
[  3] 42.0000-43.0000 sec   256 KBytes  2.10 Mbits/sec  13105/0    13107 pps
[  3] 43.0000-44.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13107 pps
[  3] 44.0000-45.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13107 pps
[  3] 45.0000-46.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13104 pps
[  3] 46.0000-47.0000 sec   256 KBytes  2.10 Mbits/sec  13109/0    13109 pps
[  3] 47.0000-48.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13108 pps
[  3] 48.0000-49.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13106 pps
[  3] 49.0000-50.0000 sec   256 KBytes  2.10 Mbits/sec  13105/0    13105 pps
[  3] 50.0000-51.0000 sec   256 KBytes  2.10 Mbits/sec  13110/0    13110 pps
[  3] 51.0000-52.0000 sec   256 KBytes  2.10 Mbits/sec  13104/0    13108 pps
[  3] 52.0000-53.0000 sec   256 KBytes  2.10 Mbits/sec  13110/0    13106 pps
[  3] 53.0000-54.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13108 pps
[  3] 54.0000-55.0000 sec   256 KBytes  2.10 Mbits/sec  13105/0    13108 pps
[  3] 55.0000-56.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13107 pps
[  3] 56.0000-57.0000 sec   256 KBytes  2.10 Mbits/sec  13108/0    13107 pps
[  3] 57.0000-58.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13106 pps
[  3] 58.0000-59.0000 sec   256 KBytes  2.10 Mbits/sec  13106/0    13106 pps
[  3] 59.0000-60.0000 sec   256 KBytes  2.10 Mbits/sec  13107/0    13108 pps
[  3] 0.0000-60.0007 sec  15.0 MBytes  2.10 Mbits/sec  786443/0    13107 pps
[  3] Sent 786443 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-60.1233 sec  7.88 MBytes  1.10 Mbits/sec   0.145 ms 373400/786443 (47%)
[  3] 0.0000-60.1233 sec  32 datagrams received out-of-order
4.1.2.2 数据简单分析
  • 测试设置为使用UDP协议,目标带宽为2Mbps,发送数据包大小为20字节,持续时间为60秒。
  • 客户端连接到IP地址为172.24.42.46的服务器。
  • 在整个测试期间,每秒发送的数据包数量保持在大约13100个左右,除非有数据包丢失,否则保持稳定。
  • 服务器报告显示,在60秒的测试期间,共发送了786443个数据包,其中约47%的数据包丢失,且有32个数据包接收时顺序错乱。
  • 带宽的实际测量结果约为1.10 Mbits/sec,明显低于设定的2Mbps目标带宽。

综上所述,这些数据表明在给定的网络环境下存在丢包和顺序错乱的问题,导致带宽表现不佳。可能的原因包括网络拥塞、传输延迟、或网络设备配置问题。为解决这些问题,可以考虑优化网络配置、增加带宽、或改善网络连接质量。

4.1.2.3 数据深入分析

本次测试已经发现一些线索,至少确实可能存在链路状况不佳的情况。

分析对比测试线A v/s B的差异:

  1. 测试参数A:默认v4 UDP=1470, v6 UDP=1450
  2. 测试参数B:数据包大小20字节,每秒报文数量13105

在维持带宽情况下,小包数量增加,会引起网络拥塞,进而导致丢包率明显增加。此时,出现花屏和卡顿就不足为奇。

从这里可以认识到,之前的测试过于简单。
因为视频流并非只是发送一种长度的报文,可能是变化的;另外,在编码完成后,会有发送报文需求;在图像编码时,视频数据尚没有准备好,此时,并无数据发送需求,网络带宽是空闲的。

引入以下思考:视频传输H264编码P/I帧的数据发送是如何分布的(时空概念)?

  1. 报文大小分布情况如何?
  2. 报文发送速率如何?
  • Bit Rate Throttling Algorithm on Video over RTP

举了一个例子:

  • 高峰时刻I帧,5ms 15个数据包,而I帧又是完整的帧图像,如果此时出现丢包,那可不就是花屏、卡顿。
  • 每间隔2秒,传输报文数量出现一个2倍峰值
  • 15FPS(66.7ms)下,H264每隔62ms出现一个峰值大包传输集中点(时刻)

在这里插入图片描述在这里插入图片描述

4.1.2.4 模拟测试网络

模拟高峰时刻带宽测试:5ms 15包数据(每秒15x1000/5=3000pps),1400字节一包。

  • 实际测试网络,还存在其他视频链路,应用程序等等网络开销
  • 模拟始终高压传输,导致带宽更加紧张,容易导致UDP丢包

最终结果:81%丢包率,压根没法看视频了(妥妥的2-8定律)

$ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 1137 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 49273 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3002 pps
[  3] 1.0000-2.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3000 pps
[  3] 2.0000-3.0000 sec  4.00 MBytes  33.6 Mbits/sec  2996/0     3000 pps
[  3] 3.0000-4.0000 sec  4.01 MBytes  33.6 Mbits/sec  3003/0     2999 pps
[  3] 4.0000-5.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3001 pps
[  3] 5.0000-6.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3000 pps
[  3] 6.0000-7.0000 sec  4.00 MBytes  33.6 Mbits/sec  2999/0     3000 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 0.0000-10.0009 sec  40.1 MBytes  33.6 Mbits/sec  30005/0     3000 pps
[  3] Sent 30005 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.3893 sec  7.64 MBytes  6.17 Mbits/sec   2.070 ms 24283/30007 (81%)

4.2 第二次测试

鉴于第一次测试的网络拓扑问题,现在只用一个4G LTE链路:因此链路独享,功能单一,不会有太多的竞争关系。【网络拓扑B】调整如下:

在这里插入图片描述

4.2.1 测试数据

$ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 1426 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 51044 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3002 pps
[  3] 1.0000-2.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 2.0000-3.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 3.0000-4.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3001 pps
[  3] 4.0000-5.0000 sec  4.00 MBytes  33.6 Mbits/sec  2999/0     2999 pps
[  3] 5.0000-6.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 6.0000-7.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 0.0000-10.0009 sec  40.1 MBytes  33.6 Mbits/sec  30005/0     3000 pps
[  3] Sent 30005 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.2656 sec  37.5 MBytes  30.7 Mbits/sec   0.268 ms 1890/30004 (6.3%)

注:更多测试数据供参考。

daniel@rpanion:~ $ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 1510 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 43161 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3002 pps
[  3] 1.0000-2.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 2.0000-3.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3001 pps
[  3] 3.0000-4.0000 sec  4.00 MBytes  33.6 Mbits/sec  2999/0     2999 pps
[  3] 4.0000-5.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 5.0000-6.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 6.0000-7.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 0.0000-10.0009 sec  40.1 MBytes  33.6 Mbits/sec  30005/0     3000 pps
[  3] Sent 30005 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0739 sec  39.7 MBytes  33.0 Mbits/sec   0.191 ms  278/30004 (0.93%)
daniel@rpanion:~ $ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 1516 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 59910 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3002 pps
[  3] 1.0000-2.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 2.0000-3.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 3.0000-4.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 4.0000-5.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 5.0000-6.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 6.0000-7.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 0.0000-10.0008 sec  40.1 MBytes  33.6 Mbits/sec  30005/0     3000 pps
[  3] Sent 30005 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.2943 sec  36.7 MBytes  29.9 Mbits/sec   0.462 ms 2551/30004 (8.5%)
daniel@rpanion:~ $ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 1520 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 40583 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3001 pps
[  3] 1.0000-2.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 2.0000-3.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 3.0000-4.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 4.0000-5.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 5.0000-6.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 6.0000-7.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 0.0000-10.0022 sec  40.1 MBytes  33.6 Mbits/sec  30009/0     3000 pps
[  3] Sent 30009 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0383 sec  38.3 MBytes  32.0 Mbits/sec   0.262 ms 1311/30008 (4.4%)
daniel@rpanion:~ $ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 1524 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 47580 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  3.96 MBytes  33.2 Mbits/sec  2967/0     3002 pps
[  3] 1.0000-2.0000 sec  4.00 MBytes  33.6 Mbits/sec  2997/0     3000 pps
[  3] 2.0000-3.0000 sec  4.03 MBytes  33.8 Mbits/sec  3015/0     2979 pps
[  3] 3.0000-4.0000 sec  4.03 MBytes  33.8 Mbits/sec  3016/0     3014 pps
[  3] 4.0000-5.0000 sec  3.98 MBytes  33.4 Mbits/sec  2982/0     3007 pps
[  3] 5.0000-6.0000 sec  4.04 MBytes  33.9 Mbits/sec  3024/0     3000 pps
[  3] 6.0000-7.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 0.0000-10.0009 sec  40.1 MBytes  33.6 Mbits/sec  30005/0     3000 pps
[  3] Sent 30005 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0151 sec  39.7 MBytes  33.3 Mbits/sec   0.302 ms  235/30004 (0.78%)
daniel@rpanion:~ $ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 1527 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 44434 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3001 pps
[  3] 1.0000-2.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 2.0000-3.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 3.0000-4.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 4.0000-5.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 5.0000-6.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 6.0000-7.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 0.0000-10.0009 sec  40.1 MBytes  33.6 Mbits/sec  30005/0     3000 pps
[  3] Sent 30005 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.0227 sec  40.1 MBytes  33.5 Mbits/sec   0.474 ms    0/30004 (0%)

4.2.2 数据分析

根据提供的iperf数据,我们可以得到以下分析:

  1. 带宽:
    平均带宽为33.6 Mbits/sec,这是UDP流量在网络上的平均传输速率。

  2. 丢包率:
    总共发送了30005个数据包,其中有1890个数据包丢失。
    丢包率为6.3%。这意味着有6.3%的数据包未能成功到达目的地。这可能是由于网络拥塞、丢包或其他网络问题引起的。

  3. Jitter:
    Jitter指的是数据包传输之间的时延变化。在本例中,Jitter为0.268毫秒。Jitter越小,表示数据包传输的稳定性越高。

  4. PPS(每秒数据包数):
    平均每秒传输了3000个数据包。这是一个可用于评估网络负载的指标。

综合来看,虽然仍存在6.3%的丢包率,实际情况不可能长时间高压力,视频码流也有FEC等纠错机制。相对于第一次测试情况,相对来说该网络链路环境更为理想,也更加符合实际使用情况。

4.2.3 网络延迟分析

通过ping命令,iperf(Linux端)和QGC(Windows端),分析网络延时:

  • iperf(Linux端):平均3.35ms
64 bytes from 172.24.161.33: icmp_seq=1 ttl=64 time=3.07 ms
64 bytes from 172.24.161.33: icmp_seq=2 ttl=64 time=3.14 ms
64 bytes from 172.24.161.33: icmp_seq=3 ttl=64 time=3.83 ms
64 bytes from 172.24.161.33: icmp_seq=4 ttl=64 time=3.35 ms
  • QGC(Windows端):平均7.75ms
Reply from 172.24.161.33: bytes=32 time=7ms TTL=64
Reply from 172.24.161.33: bytes=32 time=3ms TTL=64
Reply from 172.24.161.33: bytes=32 time=12ms TTL=64
Reply from 172.24.161.33: bytes=32 time=9ms TTL=64

4.2.4 视频效果

实测效果: ~ 245 ms 时延;???貌似VPN没啥时延啊?呵呵,后面说,卖个关子。
在这里插入图片描述

Rpanion OV5647 ZeroTier 网络拓扑B 性能实测

4.3 第三次测试

使用更加贴近实际场景,两个4G LTE独享带宽【网络拓扑C】。

在这里插入图片描述

4.3.1 测试数据

$ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 860 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 36516 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  4.01 MBytes  33.6 Mbits/sec  3002/0     3002 pps
[  3] 1.0000-2.0000 sec  4.00 MBytes  33.6 Mbits/sec  2999/0     3000 pps
[  3] 2.0000-3.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 3.0000-4.0000 sec  4.00 MBytes  33.6 Mbits/sec  2999/0     3000 pps
[  3] 4.0000-5.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3000 pps
[  3] 5.0000-6.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 6.0000-7.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     2999 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3001 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3001 pps
[  3] 0.0000-10.0015 sec  40.1 MBytes  33.6 Mbits/sec  30007/0     3000 pps
[  3] Sent 30007 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.3614 sec  37.5 MBytes  30.4 Mbits/sec   0.469 ms 1882/30006 (6.3%)
[  3] 0.0000-10.3614 sec  2921 datagrams received out-of-order

注:更多测试数据供参考。

$ iperf -b 3000pps -u -e -i 1 -l 1400 -c 172.24.42.46
------------------------------------------------------------
Client connecting to 172.24.42.46, UDP port 5001 with pid 1484 (1 flows)
Write buffer size: 1.37 KByte
UDP buffer size:  208 KByte (default)
------------------------------------------------------------
[  3] local 172.24.161.33%zt44xhvvrc port 39651 connected with 172.24.42.46 port 5001
[ ID] Interval            Transfer     Bandwidth      Write/Err  PPS
[  3] 0.0000-1.0000 sec  4.01 MBytes  33.6 Mbits/sec  3001/0     3002 pps
[  3] 1.0000-2.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 2.0000-3.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 3.0000-4.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 4.0000-5.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 5.0000-6.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 6.0000-7.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 7.0000-8.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 8.0000-9.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 9.0000-10.0000 sec  4.01 MBytes  33.6 Mbits/sec  3000/0     3000 pps
[  3] 0.0000-10.0005 sec  40.1 MBytes  33.6 Mbits/sec  30004/0     3000 pps
[  3] Sent 30004 datagrams
[  3] Server Report:
[ ID] Interval       Transfer     Bandwidth        Jitter   Lost/Total Datagrams
[  3] 0.0000-10.1356 sec  36.8 MBytes  30.4 Mbits/sec   0.174 ms 2446/30003 (8.2%)

4.3.2 数据分析

  1. 测试参数:
    使用了以下参数进行测试:带宽设置为3000pps、使用UDP协议、启用错误报告、设置报告间隔为1秒、每个数据包大小为1400字节,连接到IP地址为172.24.42.46的服务器。

  2. 客户端信息:
    客户端IP地址为172.24.161.33,端口36516,连接到服务器IP地址为172.24.42.46,端口5001。

  3. 传输数据:
    在每个1秒的时间间隔内,传输了大约4.01兆字节的数据。
    带宽在每秒都维持在约33.6兆位/秒。
    平均每秒发送的数据包数(PPS)为3000左右。

  4. UDP错误:
    未发现UDP错误。

  5. 服务器报告:
    在10.36秒的测试期间,总共传输了37.5兆字节的数据。
    平均带宽为30.4兆位/秒。
    报告的抖动(jitter)为0.469毫秒。
    6.3%的数据包丢失率,共丢失1882个数据包。
    有2921个数据包收到时是乱序的。

综上所述,该iperf测试表明UDP连接在设定的条件下运行良好,但存在一定程度的数据包丢失和乱序。

4.3.3 网络延迟分析

通过ping命令,iperf(Linux端)和QGC(Windows端),分析网络延时:

  • iperf(Linux端):平均607.5ms
64 bytes from 172.24.161.33: icmp_seq=1 ttl=64 time=668 ms
64 bytes from 172.24.161.33: icmp_seq=2 ttl=64 time=583 ms
64 bytes from 172.24.161.33: icmp_seq=4 ttl=64 time=565 ms
64 bytes from 172.24.161.33: icmp_seq=5 ttl=64 time=614 ms
  • QGC(Windows端):平均822.25ms
Reply from 172.24.161.33: bytes=32 time=825ms TTL=64
Reply from 172.24.161.33: bytes=32 time=854ms TTL=64
Reply from 172.24.161.33: bytes=32 time=801ms TTL=64
Reply from 172.24.161.33: bytes=32 time=809ms TTL=64

4.3.4 视频效果

实测效果: ~ 569 ms 时延; Linux端607.5/2 = 303ms, 569 - 303 = 266 ms左右。

在这里插入图片描述

Rpanion OV5647 ZeroTier 网络拓扑C 性能实测

5. 总结

综上所述:

  1. 【网络拓扑A】~ 200 ms, 参考:《ArduPilot开源飞控之lida2003套机+伴机电脑首飞》
  2. 【网络拓扑B】~ 245 ms
  3. 【网络拓扑C】~ 569 ms
  4. 局域网点对点性能最好,直接就是Rpanion+Rpi3B+ 200ms时延。
  5. 统一WLAN出口,VPN仍然能够很好的优化,与局域网性能相当。
  6. 实际不同发射/接受链路网络,若VPN不做好优化,时延会很显著。

实际使用场景,需要看链路情况,若要进行商业部署,必须做好优化,或者自建网络(VPN/P2P等等)来控制好链路时延问题。

Ardupilot + Rpi3B+ + OV5647 + 4G LTE Copter

6. 参考资料

【1】zerotier - Securely connect any device, anywhere.

7. ZeroTier优化办法 - Moon节点

7.1 服务端配置

  • Step1:安装zerotier-one
# curl -s https://install.zerotier.com/ | sudo bash
  • Step2:加入自己的网络
# zerotier-cli join <Network ID>
  • Step3:生成moon.json文件
# cd  /var/lib/zerotier-one
# zerotier-idtool initmoon identity.public > moon.json
  • Step4:修改moon.json文件 //服务器,打开安全组,开放9993端口,采用UDP协议
# vi moon.json  //字段中修改:“stableEndpoints”: [“服务器的ip/9993”]
  • Step5:生成签名文件 //生成一个000000xxxxxxxxx.moon的文件
# zerotier-idtool genmoon moon.json
  • Step6:新建自己的moon节点配置
# mkdir moons.d
# mv 000000xxxxxxxxx.moon moons.d
# service zerotier-one restart  //之后重启服务器端B的Zerotier

7.2 客户端配置

  • Step1:下载配置文件
# scp usrname@ServerIP:/var/lib/zerotier-one/moons.d/000000xxxxxxxxx.moon .
  • Step2:前往zerotier安装目录 //window一般在 C:\ProgramData\ZeroTier\One
  • Step3:新建ZeroTier\One\moons.d目录,并复制00000xxxx.moon
  • Step4:重启windows服务
    在这里插入图片描述- Step5:查看moon服务器 //确认服务器(IP)和节点已经在列表中,表示成功
# zerotier-cli listpeers
  • Step6:查看ping延迟 //网络跳转越少,路线越短,性能越好。国内典型值在20ms左右。

在这里插入图片描述

8. DJI RC系统性能

DJI RC Pro features O3+ video transmission technology, capable of sending a 1080p/60fps live feed from up to 15 km away and latency as low as 120 ms.

在这里插入图片描述

本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如若转载,请注明出处:http://www.rhkb.cn/news/324422.html

如若内容造成侵权/违法违规/事实不符,请联系长河编程网进行投诉反馈email:809451989@qq.com,一经查实,立即删除!

相关文章

【八】centos7.6安装chrome和chromedriver并启动selenium

学习来源&#xff1a; 安装chrome和chrome driver -----https://blog.csdn.net/zdlcome/article/details/133813441 安装Python11 -----https://blog.csdn.net/weixin_43741408/article/details/130251102 chromedriver下载地址 -----https://googlechromelabs.github.io/chrom…

# 从浅入深 学习 SpringCloud 微服务架构(八)Sentinel(1)

从浅入深 学习 SpringCloud 微服务架构&#xff08;八&#xff09;Sentinel&#xff08;1&#xff09; 一、sentinel&#xff1a;概述 1、前言 – 服务熔断 Hystrix 的替换方案。 1&#xff09;2018年底 Netflix 官方宣布 Hystrix 已经足够稳定&#xff0c;不再积极开发 Hys…

用标准的GNU/Linux命令替换Alpine上的精简版命令

Alpine Linux 是一个基于 musl libc 和 busybox 的轻量级Linux发行版&#xff0c;busybox 实现了很多常用类Unix命令的精简版&#xff0c;特点是体积很小&#xff0c;舍弃了很多不常用参数&#xff0c;我们简单对比一下标准Linux自带的 date 命令 和 Alpine下默认的 date 命令便…

初期Linux

一&#xff0c;系统分为 1.1window系统 个人 &#xff1a;win7&#xff0c;win8&#xff0c;Win10&#xff0c;Win11服务器版&#xff1a;window server 2003&#xff0c;window server 2008 1.2Linux系统 centos7redhatubantukali 1.3什么是Linux&#xff1f; Linux是基…

连通“数据”,让制造变“聪明”

说起数据智能&#xff0c;你第一时间想到的是什么呢&#xff1f;是科技感十足的智慧城市&#xff1f;还是炫酷的人工智能景象&#xff1f; 数据作为企业的战略资产越来越受到重视&#xff0c;从最初的数据协助业务协同&#xff0c;转化为数据驱动业务&#xff0c;数据驱动运营…

ORACLE ODAX9-2的一个误告警Affects: /SYS/MB的分析处理

在运维的多套ORACLE ODAX9-2版本&#xff0c;都遇到了一个计算节点的告警&#xff1a;Description: The service Processor poweron selftest has deteced a problem. Probabity;:100, UulD:cd1ebbdf-f099-61de-ca44-ef646defe034, Resource:/SYS/MB,&#xff1b;此告警从描述上…

静态分析-RIPS-源码解析记录-01

token流扫描重构部分&#xff0c;这一部分主要利用php的token解析api解析出来的token流&#xff0c;对其中的特定token进行删除、替换、对于特定的语法结构进行重构&#xff0c;保持php语法结构上的一致性 解析主要在lib/scanner.php中通过Tokenizer这个类来实现,也就是在main…

怎么把手机ip地址变成了外省

在日常使用中&#xff0c;有时我们可能因为某些原因需要快速切换手机的IP地址&#xff0c;特别是当需要从一个省份切换到另一个省份的IP时。这种需求可能来源于网络访问限制、地理位置相关服务的使用、或者网络安全等方面的考虑。那么&#xff0c;怎么把手机IP地址变成外省呢&a…

测评工作室的养号成本,效率,纯净度,便捷性等问题怎么解决?

大家好&#xff0c;我是南哥聊跨境&#xff0c;最近有很多做测评工作室的朋友找到南哥&#xff0c;问我有什么新的测评养号系统可以解决成本&#xff0c;效率&#xff0c;纯净度&#xff0c;便捷性等问题 测评养号系统从最早的模拟器、虚拟机到911、VPS、手机设备等&#xff0…

正点原子[第二期]Linux之ARM(MX6U)裸机篇学习笔记-12-蜂鸣器

前言&#xff1a; 本文是根据哔哩哔哩网站上“正点原子[第二期]Linux之ARM&#xff08;MX6U&#xff09;裸机篇”视频的学习笔记&#xff0c;在这里会记录下正点原子 I.MX6ULL 开发板的配套视频教程所作的实验和学习笔记内容。本文大量引用了正点原子教学视频和链接中的内容。…

【大模型】LLaMA-1 模型介绍

文章目录 一、背景介绍二、模型介绍2.1 模型结构2.2 模型超参数2.3 SwiGLU 三、代码分析3.1 模型结构代码3.2 FairScale库介绍 四、LLaMA家族模型4.1 Alpaca4.2 Vicuna4.3 Koala(考拉)4.4 Baize (白泽)4.5 Luotuo (骆驼&#xff0c;Chinese)4.6 其他 参考资料 LLaMA&#xff08…

Human β-NGF ELISA试剂盒

走近β-NGF 神经生长因子(nerve growth factor, NGF)最初从小鼠颌下腺中以7S复合体的形式分离而得&#xff0c;复合体由三个非共价连接的亚基α,β和γ组成。 NGF的α和β亚基均属于丝氨酸蛋白酶组织激肽释放酶家族成员&#xff0c;β亚基也称为β-NGF或2…

国产银河麒麟V10SP1系统下搭建TiDB数据库操作步骤图文

开发目的&#xff1a;在国产银河麒麟系统中搭建TiDB数据库运行环境。 开发工具&#xff1a;银河麒麟系统V10SP1TiDBMySql数据库8.0。 具体步骤&#xff1a; 1、在VmWare虚拟机中安装好国产银河麒麟V10Sp1操作系统。 2、打开终端命令&#xff0c;安装TiDB相关软件&#xff1…

ALV Color-颜色

目录 前言 实战 列颜色 行颜色 单元格颜色 前言 在ABAP ALV中&#xff0c;Color颜色设置是一种增强列表显示效果的重要手段&#xff0c;可以用来突出显示特定行、列或单元格&#xff0c;以吸引用户注意或传达数据的特定状态。 颜色设置中有优先级顺序&#xff0c;他们是单元格…

线上剧本杀小程序:为行业带来新的活力,未来可期

剧本杀是一项新型的社交游戏活动&#xff0c;从前几年开始就呈现了快速发展态势&#xff0c;为大众带来沉浸式的游戏体验&#xff0c;一度成为年轻人娱乐休闲消费的首选方式&#xff0c;吸引了大量的消费者和商家。 不过&#xff0c;在市场发展中&#xff0c;剧本杀行业仍需要…

掌握文件重命名技巧:一次性处理多路径文件并赋予独立编号

在日常工作和生活中&#xff0c;我们经常需要处理大量的文件&#xff0c;而文件重命名则是一项非常常见的任务。如何高效地一次性处理多路径文件并赋予独立编号&#xff0c;成为许多用户关注的焦点。本文将介绍云炫文件管理器一些实用的文件重命名技巧&#xff0c;帮助您轻松应…

基于FPGA的去雾算法

去雾算法的原理是基于图像去模糊的原理&#xff0c;通过对图像中的散射光进行估计和去除来消除图像中的雾霾效果。 去雾算法通常分为以下几个步骤&#xff1a; 1. 导引滤波&#xff1a;首先使用导引滤波器对图像进行滤波&#xff0c;目的是估计图像中散射光的强度。导引滤波器…

《这就是ChatGPT》读书笔记

书名&#xff1a;这就是ChatGPT 作者&#xff1a;[美] 斯蒂芬沃尔弗拉姆&#xff08;Stephen Wolfram&#xff09; ChatGPT在做什么&#xff1f; ChatGPT可以生成类似于人类书写的文本&#xff0c;它基本任务是弄清楚如何针对它得到的任何文本产生“合理的延续”。当ChatGPT写…

Spring框架学习笔记(一):Spring基本介绍(包含IOC容器底层结构)

1 官方资料 1.1 官网 https://spring.io/ 1.2 进入 Spring5 下拉 projects, 进入 Spring Framework 进入 Spring5 的 github 1.3 在maven项目中导入依赖 <dependencies><!--加入spring开发的基本包--><dependency><groupId>org.springframework<…

STC -PWM

一.STC8H1K16初始化,以下一步配置后就会有波形输出. // // 函数: PWMB_Output_init // 描述: 用户初始化程序. // 参数: None. // 返回: None. // 版本: V1.0, 2020-09-28 //u16 PWM8__setDuty25000;u16 PWM8__setPeriod50000; void PWMB_Output_init(void) {PWMx_InitDefi…