Regular GNSS receivers use scalar tracking loops to track each satellite signal in view independently. In challenging environments, like urban canyons, the visibility of the satellite signal is not always good enough to reach a stable tracking. Signal degradations have a high impact and reduce the robustness and precision of the receiver tracking and position, velocity, and time (PVT) solution. Another approach, the vector tracking (VT), where signals are tracked with relation to each other, overcomes this weakness. There, the tracking and PVT solution estimation is done with an interlaced algorithm. Additionally, the tracked satellites are connected via an extended Kalman Filter. Present developments have shown good results in a light urban environment with a real time implementation of the vector delay lock loop (VDLL). There, only the scalar code tracking loop was replaced with a vector approach. The carrier loops are still independent and closed with scalar tracking loops. To achieve even better results this paper explains the enhancement of the VDLL to a vector delay and frequency lock loop (VDFLL), where the complete scalar tracking is replaced by the vector loop. A simulated test with GPS L1 C/A signals was done and the results are presented and discussed. With the VDFLL the system has to deal with additional error and noise due to the included carrier tracking. It requires an appropriate system model for effective operation, and the loops have to be closed simultaneously for all channels and as fast as possible to avoid timing issues.