Fifth-generation (5G) networks provide high-speed data transmission service. Therefore, relying on effective congestion control methods is one of the secrets to the success of the network's performance. One of the key functions of Transmission control protocol (TCP) is congestion control, which attempts to limit the TCP connection to its fair share of network bandwidth. TCP should have the ability to adapt to network conditions and channel fluctuations to get highly reliable and efficient data communications. This paper presents an adjusted fast recovery algorithm to enhance the performance of TCP-Newreno. The main idea behind this modification is to adjust the congestion window (cwnd) based on evaluating the round-trip time (RTT) changes to adapt to the dynamic channel conditions rather than reducing it to half of its value when network congestion is detected. In 5G networks, channel conditions tend to change frequently. This adjustment increases network throughput, reduces packet loss, and decreases delay.