Downlink Failed: RX_Timeout

Hi, Iam using RFM95 as lorawan transmitter to the RAK Lorawan Gateway. Uplink data is successfully transmitted, but not with downlink data, apparently it has issue with RX_Timeout that keep appear after successfully transmitted an uplink Data. I tried send downlink data to node but failed due to this Issue.
here the debug result:

```
Mbed LoRaWANStack initialized

CONFIRMED message retries : 3

set class OK

Adaptive data rate (ADR) - Disable

set datarate OK

Connection - In Progress …

Connection - Successful
[2K[39m[INFO][LMAC]: RTS = 14 bytes, PEND = 0, Port: 6[0m

14 bytes scheduled for transmission
[2K[31m[ERR ][LMACC]: Invalid MAC command (0x0)![0m
[2K[31m[ERR ][LSTK]: RX_ERROR: mcps_indication status = 1[0m

Message Sent to Network Server

Error in reception - Code = 9
[2K[39m[INFO][LMAC]: RTS = 14 bytes, PEND = 0, Port: 6[0m

14 bytes scheduled for transmission
[2K[31m[ERR ][LMACC]: Invalid MAC command (0x0)![0m
[2K[31m[ERR ][LSTK]: RX_ERROR: mcps_indication status = 1[0m

Message Sent to Network Server

Error in reception - Code = 9
```

Error no 9: RX_ Timeout
mcps_indication status no 1: LORAMAC_EVENT_INFO_STATUS_ERROR

IDE : Mbed IDE
OS: Version 6.10
Lorawan server: Chirpstack,
Gateway: RAK with mqtt bridge V.2.7.1

I've compared with another hardware e.g RAK811 (default firmware), result is OK, so I Assume there is nothing wrong with Chirpstack and RAK Gateway
Any solution would be appreciated
Thanks

The report identifies the Mbed LoRaWAN stack, RFM95, ChirpStack, and the RAK gateway, but names no source file or test. Start by reproducing the uplink/downlink exchange and tracing the RX_Timeout and invalid MAC command statuses, then compare the working RAK811 setup. Done means the cause is documented and downlink reception is verified.

Written by the indexing model from the issue text.