Experience up to 1000x Goodput Improvement with TCP M-AT
TCP M-Adaptive Transmission (TCP M-AT) significantly enhances network performance. Compared to existing protocols like TCP Cubic and TCP Google BBR, TCP M-AT offers a remarkable up to 1000x increase in throughput/ goodput. This leads to faster data transfer rates and more efficient use of network resources, ultimately improving user experience and reducing operational costs.
REAL-WORLD TEST RESULTS (LINUX KERNEL 6.17.0 OVER POOR WIFI)
Test Component | Specification / Description |
|---|---|
| Operating System | Linux (Kernel version 6.17.0) |
| Network Type | Wi-Fi link, intentionally degraded with high packet loss and fluctuating RTT |
| CCA Tested | TCP Cubic, TCP Google BBR, TCP M-Adaptive Transmission (TCP M-AT) |
| Measured Metric | Average Throughput (KB/s) |
| Packet Loss / Noise | Real-world, variable wireless interference |
| Test Methodology | Simultaneous upload throughput tests using an identical system setup |
Β
| Congestion Control Algorithm | Average Throughput (KB/s) | Upload Cumulative Progress |
|---|---|---|
| TCP Cubic | ~17 KB/s | 0.81% |
| TCP Google BBR | ~2.2 KB/s | 0.19% |
| TCP M-AT | ~1740 KB/s | 100% |
YouTube Demo Link: https://youtu.be/EypR2cUPubc?si=_ipC-AzvcmA3HrYt
Key Takeaway:
Under identical, adverse wireless conditions TCP M-AT v2.0 averaged 1.7 MB/s, compared with ~2.2 KB/s (Google BBR) and ~17 KB/s (Cubic) β i.e., β773Γ the throughput of Google BBR (β77,173% improvement) and β100Γ the throughput of Cubic (β10,000% improvement). The connection also reached ~1.9 MB/s for a sustained interval, which corresponds to β864Γ Google BBR (~86,264% improvement) and β114Γ Cubic (~11,400% improvement). These results demonstrate that M-ATβs adaptive RTT evaluation, refined bandwidth memory factor, precise real-time bandwidth estimation, and proactive congestion avoidance keep server transmissions stable and high-throughput when conventional algorithms collapse. Practically, this means a one-time server/edge deployment can unlock substantial performance gains for all connected clients β making M-AT a high-leverage solution for cloud providers, telcos, IoT platforms and satellite operators seeking predictable, efficient data transfer without per-device changes.Β
Β
SIMULATION-BASED TESTING:
| Network Environment | TCP M-AT | Google TCP BBR v3 | TCP Cubic | Improvement Over BBR | Improvement Over Cubic |
|---|---|---|---|---|---|
| LTE (10% packet loss) | 9.5 Mbps | 6 Mbps | 1.2 Mbps | ~58% | ~692% |
| 5G (5% packet loss) | 28β30 Mbps | 20 Mbps | 3 Mbps | ~40β50% | ~833β900% |
| LEO Satellite (5% loss) | 30 Mbps | 18 Mbps | 2.5 Mbps | ~65% | ~1200% |
β Scroll to view more β
Β
Cloud Computing Optimization
TCP M-AT significantly improves data transfer rates in cloud environments, reducing latency and enhancing the performance of cloud-based applications and services.
Enhanced 5G Network Performance
By optimizing data transmission over 5G networks, TCP M-AT ensures faster download and upload speeds, providing a seamless user experience for mobile applications.
IoT Data Transmission Efficiency
TCP M-AT facilitates reliable and efficient data transmission for IoT devices, enabling real-time monitoring and control in smart homes, industrial automation, and more.
Satellite Communication Enhancement
TCP M-AT optimizes bandwidth usage in satellite communication, ensuring stable and high-speed data transfer for remote locations and maritime applications.