Toolbox for simulation of transmission of SPIHT encoded images over unreliable channels, for instance wireless or packet-erasure channels. Requires Zerotree Coding and Quality Assessment toolboxes, and the MATLAB Communications Toolbox. Note that the toolbox contains more functions than that is listed here below, I yet have to check those for consistency before putting them on the list. Some of the functions are still missing an adequate help info, this will be done as soon my time allows. Please contact me directly if you need some info sooner than that.
Functions
This toolbox consists of the following functions:
- BSC_BER - Simulation of the Binary Symemtric Channel (BSC)
- awgn_EsN0 - Simulation of the Additive White Gaussian Noise (AWGN) channel
- GE_awgn - Simulation of the Gilbert-Elliot (GE) channel, using two AWGN channels
- get_channel - Loads a predefined channel (saved in a txt file)
- get_RCPC_code - Loads a predefined RCPC code from a default code family ('Punct_codes.txt')
- RCPC_encode - Rate Compatible Punctured Convolutional (RCPC) coding of the binary stream
- RCPC_test_equal_data - Transmission of packets with constant data part, using RCPC
- RCPC_test_equal_packet - Transmission of packets with constant size, using RCPC
- optimal_RCPC_equal - Computes the Equal Error Protection (EEP) scheme for a given image and channel
- send_image_equal - Transmission of SPIHT encoded image using EEP
- send_image_equal_RS - Transmission of SPIHT encoded image using EEP and Reed-Solomon protection
- optimal_RCPC_unequal - Computes the Unequal Error Protection (UEP) scheme for a given image and channel
- send_image_unequal - Transmission of SPIHT encoded image using UEP
There are also some scripts to measure the channel error rate, RCPC code characteristics and Rate-Distortion (R-D) curve of the SPIHT compression on the particular image:
- script_channel_performance - Measures the BER and PER values for RCPC transmission on a given channel
- RDcurveBytePrec - Runs SPIHT over range of target bitrates and stores the R-D curve
Examples
get_channel (contents of the file 'AWGN1.txt'):
ch_handle=@awgn_EsN0; parametar.EsN0=-0.8556; parametar.state=-1; parametar.Bch=128000; parametar.Nch=256;
get_RCPC_code (one RCPC entry from the file 'Punct_codes.txt'):
8/9 [Hag88] Memory=6 CodeGenerator=[133 171 145] PunctCode=f78800 cd=[0,0,24,740,13321,217761,3315491,48278177] %Punctcode=[f7,88,00]=[1 1 1 1 0 1 1 1,1 0 0 0 1 0 0 0,0 0 0 0 0 0 0 0]
get_RCPC_code, RCPC_encode:
>> signal=round(rand(1,8))
signal =
0 1 1 0 0 0 1 0
>> [Memory,Ib,Kb,t,PN,P,TotRate,PunctInd] = get_RCPC_code('8/9');
>> outsignal = RCPC_encode(signal,Memory,t,P,PunctInd)
outsignal =
1 1 -1 -1 -1 1 -1 1 1
RCPC_test_equal_data:
>> [diffvec,BER,PER]=RCPC_test_equal_data(100,128,'AWGN1','1/2',8,0); Mother code rate: 1/3 Puncturing code rate: 16/24 Total rate: 1/2 Packet size: 284 Number of packets: 100 Transmitted data bits: 12800 ***Errors Statistics*** Errors = 52 BER = 0.004063 Packet errors = 12 PER = 0.120000 (CRC) Correctly detected packet errors = 12 (CRC) Detected packet errors = 12
script_channel_performance:
>> script_channel_performance(@RCPC_test_equal_packet,100000,16,'GE1');
optimal_RCPC_equal:
>> optimal_RCPC_equal(1000,16,'GE1','Lena512RD','draw_small'); RCPC total rate: 1/1 - average PSNR: 17.281731 RCPC total rate: 8/9 - average PSNR: 18.406105 RCPC total rate: 4/5 - average PSNR: 18.432578 RCPC total rate: 2/3 - average PSNR: 19.129653 RCPC total rate: 4/7 - average PSNR: 21.051831 RCPC total rate: 1/2 - average PSNR: 24.052347 RCPC total rate: 4/9 - average PSNR: 26.976499 RCPC total rate: 2/5 - average PSNR: 29.522427 RCPC total rate: 4/11 - average PSNR: 30.491567 RCPC total rate: 1/3 - average PSNR: 30.432119 *** Minimal distortion RCPC code: 4/11 PSNR: 30.491567
