LoPy Nano-Gateway

[FelixDonkers]

@gertjanvanhethof I'm also interrested to use LoPy as a LoRaWan gateway. Looking forward to a solution.

[Colateral]

@Roberto We setup a test bench with 3 LoPy following the idea of your code. We did the test with blocking socket and thread and non blocking sockets. The result is the same.

2 are the nodes configured as tx_iq and and one is the Gw (rx_iq).

The script is simple: each device (node and GW) is sending 29 bytes message size (_MSG_FORMAT = "!HHLBB%ds") each 2 seconds. In the msg data we are printing a counter (that varies from device). That counter is also in the header (see above)

We noticed that sometimes if you start the gw after the nodes... randomly some of the nodes are not receiving the data, despite the fact the GW is receiving the data from the nodes. And we observed also that sometimes when we start the node after the GW, the GW is not receiving the data from the nodes.

You need to reset the node or GW in order to get the broadcasted data from to the others... and this hazardous and not at all good in real deployment.

BUT THE MOST PROBLEMATIC is the fact that the 29bytes are partially received, sometimes you are getting 15bytes or whatever number less than 29bytes.

Have you experience this connection issue related with nodes/gw starting order?
Does LoPy stack is guarantee the receiving of msg in one chunk? ( if you send 29bytes you got 29bytes)

Here are some samples:
good msg receive

from: 800
to: 0
Id: 3058
Flags: 0
PayloadSize: 18
Payload: b'3058- 1234567890 '
<<< beat msgId 3066
bufferSize 29
pkgSize 28
data_size 18
remaining buffer 0

bad msg receive

from: 800
to: 0
Id: 3059
Flags: 0
PayloadSize: 18
Payload: b'3059- 1234567890 '
bufferSize 15
buffer too small <<< ustruct exception

[Roberto]

Hi @Colateral,

Yes, i have noticed that sometimes some messages arrive malformed or with bytes missing. This could be due to LoRa message collision between devices. If two devices send at the same time in the same band this can happen.

There is a new blog here in the forum with a new code suggestion that includes.

1. Message length check on the nano-gateway
2. Message retry on the nodes
3. Max timeout in the nodes

Number 3 will help with the fact that you have to have the nano-gateway on before you start the nodes. In this code, after sending a message the node waits in an infinite loop for the ack. The problem with this approach is that if no ack is received (the nano-gateway was not connected yet) the node will stay in that loop until reseted.

Keep in ming that both this and the new code are just examples for specific usages and they require adaptation depending on your needs.
The link for the new post is LoPy Nano-Gateway Extended (Timeout and Retry)

Best,
Roberto

[feabhas]

@Roberto I have successfully got a 2 node and 1 gateway setup working based on your code, thanks very much. One minor query, In the gateway you use

_LORA_PKG_FORMAT = "!BB%ds"

but in the node_1.py you miss off the !, e.g.

_LORA_PKG_FORMAT = "BB%ds"

Shouldn't they be consistent?

[Roberto]

Hi @feabhas
yeah, that sounds about right. Change that please. Ill edit the post accordingly Thanks

[Colateral]

@Roberto Your example it is a good start but looping and blocking might not be a good approach.

We built another script that is using threads with blocking socket ... and run the script on a minimal Lora "star" network: one GW and 2 NODES. For each device you have to configure tx,rx and device id in cfg file . and upload the script and cfg file on it and run the script in Putty (not Pymakr). We didn't faced with this script clashing issues.. but we faced other issues.

You will see that after awhile on GW, the script is ending with a memory leak... but this is never happening on the nodes.
If you run the script from 3 Pymakr instances ... sometimes is working for hours and after that the GW is stop sending.

We conclude that is something wrong on the stack.

The issue might be related only to rx_iq configuration.
https://drive.google.com/drive/folders/0BzRE8IpnMMnLNHBseDJWblN4VGM?usp=sharing

[Roberto]

Hi @Colateral
As i wrote in my last message, there is a new example of the nano-gateway with a non-blocking loop approach.
Regarding the memory issues we will look into this. I left the code on the other post running for 3 hours without any problems. I will leave it running today on a test bench for 24+ h to see if there are any problems and monitor the memory while it transmits. Will let you know the results.

The link to the new code is
https://forum.pycom.io/post/3242

[Colateral]

@Roberto HI Roberto. Thanks. I ran the last script and is running fine. Regarding the memory leak, I believe is related with socket blocking approach. If we change the script and move to non blocking (as your script does) everything works fine.

[Roberto]

@Colateral
Thanks for the quick response, i'll try a blocking approach to see if i can reproduce the error and check whats the root of it.

[assia]

@Colateral said in LoPy Nano-Gateway:
Hi @Roberto i am bbegginer in LoRa and i don't understand the use of rx_iq and tx_iq.!
thanks

[borjam]

@assia said in LoPy Nano-Gateway:

@Colateral said in LoPy Nano-Gateway:
Hi @Roberto i am bbegginer in LoRa and i don't understand the use of rx_iq and tx_iq.!

It's easy. A LoRaWAN network is not a peer to peer network. There is a master (the gateway) that communicates with the nodes, but there is no node to node communication.

Imagine that one of your nodes has the receiver active and another one transmits. The transmission is intended to be received by the gateway, of course (we said that there is no node-to-node communication) but the receiving node will get the packet anyway, if only to discard it.

That reception and discarding has an energy cost. How do you avoid it? By having the nodes transmit in a mode that a node can't understand, only the gateway.

Now, imagine this scenario: you make the nodes use tx_iq inversion (which is a sort of signal inversion) and the gateway use rx_iq inversion.

In this case, a signal transmitted by a node will be inverted, and the gateway will successfully decode it. But other nodes, expecting a non inverted signal, won't detect the inverted signal. So you have just avoided the unintended packet reception problem.

[assia]

@borjam
Thanks for the quick response, It's much clearer now.

[Colateral]

@borjam So how you can implement a mesh network using Lora? You put the GW in both tx and rx ?

[assia]

Hi @Roberto I tried to modify your code because I need to send a payload that contains the current time with a header containing mac address of the node and the length of the payload. Everything went well on the node side but rated nano gateway I have an Index Error: bytes index out of rang. it is because the adresse mac ? knowing that lora.mac() give : b'p\xb3\xd5I\x95kA\xf3' with length=8 but i don't understand what does mean the : b'p......' and how the length is 8 !!

thanks

node

# LoRa node

SRC_ADR_MAC=lora.mac()
#compteur packets 
PACKETs_Cnt=5
while(PACKETs_Cnt>0):
   
    # Package send containing a simple string
    msg=time.time()
    msg_len=len(str(msg))
    
    pkg = struct.pack('!%dsBi' % len(SRC_ADR_MAC), SRC_ADR_MAC, msg_len, msg)
    
    lora_sock.send(pkg)
    
    PACKETs_Cnt=PACKETs_Cnt-1
    time.sleep(5)

nano GW

# LoRa GW

while (True):
   
    recv_pkg = lora_sock.recv(255)
    if (len(recv_pkg) > 2):

        device_adr_mac, msg_len, msg = struct.unpack('!%dsBi' %8 ,   recv_pkg) 
# %8 because len(lora.mac())=8

        print('Device: %s - Pkg:  %s' % (device_adr_mac, msg))

    time.sleep(5)

[MJ]

Hi there, i am still new to lopy. I have try the example of the Lora.nanogateway code (https://docs.pycom.io/chapter/tutorials/lopy/lora-mac-nano-gateway.html) with only 1 lopy as gateway and 1 lopy as node. There is no response between both of them after i have coded it. What happen to this? Thank you.

[Colateral]

@mj Are you using 2 instances of pymaker ? If not, describe what you did.

[MJ]

@colateral Hi, i am using 1 gateway that only with pymaker opened while the node i using the power bank to power up. I have try both in vice versa yet no result showing out.